Proc. Natl. Acad. Sci.
USA
Vol. 93, pp. 2764-2769, April 1996
Ecology
Environmental services of biodiversity
NORMAN MYERS
Green College, Oxford University, United Kingdom OX3 8SZ
Contributed by Norman Myers, September 29, 1995
ABSTRACT Humans derive many utilitarian benefits
from the environmental services of biotas and ecosystems.
This is often advanced as a prime argument to support
conservation of biodiversity. There is much to be said for this
viewpoint, as is documented in this paper through a summary
assessment of several categories of environmental services,
including regulation of climate and biogeochemical cycles,
hydrological functions, soil protection, crop pollination, pest
control, recreation and ecotourism, and a number of miscel-
laneous services. It is shown that the services are indeed
significant, whether in ecological or economic senses. Partic-
ularly important is the factor of ecosystem resilience, which
appears to underpin many of the services. It should not be
supposed, however, that environmental services stem necessarily
and exclusively from biodiversity. While biodiversity often plays
a key role, the services can also derive from biomass and other
attributes of biotas. The paper concludes with a brief overview
assessment of economic values at issue and an appraisal of the
implications for conservation planning.
I. Introduction
Conservation biologists increasingly face the question, What is
biodiversity good for? Naive as this may seem to some, it is a
valid question. There is no longer enough room for a complete
stock of biodiversity on an overcrowded planet with almost six
billion humans and their multifarious activities, let alone a
projected doubling of human numbers and a tripling or
quadrupling of salient activities. So biodiversity must stake its
claims for living space in competition with other causes.
Generally speaking, biodiversity must urge the merits of its
cause through what it contributes to human welfare, preferably
doing it in the way that most appeals to political leaders and
the general public, namely in economic terms.
In response to the question above, conservation biologists
proclaim the many significant contributions of biodiversity to
the human cause. There are two categories of contributions:
material goods and environmental services. The first has been
frequently and widely documented (1-4), principally in the
form of new and improved foods, medicines and drugs, raw
materials for industry, and sources of bioenergy. The second
has been far less documented even though the issue was
identified as unusually significant almost two decades ago (5)
and even though its total value is surely far greater than that
of the first (1, 6-9). The main reasons for this lamentable
lacuna are that scientists find it much harder to demonstrate
the precise nature of the services, and it is still harder to
quantify them economically. Whereas the benefits of material
goods tend to accrue to individuals, often as producers or
consumers in the marketplace, the values of environmental
services generally pertain to society, and hence they mostly
remain unmarketed (10, 11).
This paper reviews our knowledge and understanding of the
principal services at issue. The services are extremely diverse,
The publication costs of this article were defrayed in part by page charge
payment. This article must therefore be hereby marked "advertisement" in
accordance with 18 U.S.C. §1734 solely to indicate this fact.
2764
and they occur in every last segment of the biosphere. So the
paper is perforce restricted to an illustrative selection of the
more significant services.
II. Conceptual Background
Biodiversity embraces the totality of life forms, from the
planetary spectrum of species to subunits of species (races,
populations) together with ecosystems and their ecological
processes. The species component includes all plants, animals,
and microorganisms, of which there are between 8 million and
30 million (conceivably 100 million) (12). The subspecies
component includes populations, ofwhich there could be many
billions (13). Spanning both these main categories are various
subdivisions, including community diversity, food web diver-
sity, keystone diversity, and functional diversity.
Environmental services are also known as ecosystem ser-
vices,* both terms reflecting environmental functions and
ecological processes. They can be defined as any functional
attribute of natural ecosystems that are demonstrably benefi-
cial to humankind (15). They comprise the main indirect values
of biodiversity, as opposed to direct values in the form of
material goods such as timber, fish, plant-based pharmaceu-
ticals, and germ-plasm infusions for major crops. They include
generating and maintaining soils, converting solar energy into
plant tissue, sustaining hydrological cycles, storing and cycling
essential nutrients (notably in the form of nitrogen fixation),
supplying clean air and water, absorbing and detoxifying
pollutants, decomposing wastes, pollinating crops and other
plants, controlling pests, running biogeochemical cycles (of
such vital elements as carbon, nitrogen, phosphorus, and
sulfur), controlling the gaseous mixture of the atmosphere
(which helps to determine climate), and regulating weather
and climate at both macro and micro levels. Thus they basically
include three forms of processing, namely of minerals, energy,
and water (16). In addition, biodiversity provides sites for
research, recreation, tourism, and inspiration (1, 17, 18).
The bulk of this paper will be given over to describing and
evaluating certain of these services. But first, a couple of
caveats. It is far from true that all forms of biodiversity can
contribute all environmental services or that similar forms of
biodiversity can perform similar tasks with similar efficiency.
How far do environmental services depend upon biodiversity
per se? Recent research suggests that they are highly resilient
to some loss of species and they can keep on supplying their
services even in highly modified states. A sugarcane plantation
may be more efficient at producing organic material than the
natural vegetation it replaced, and a tree farm may be more
capable of fixing atmospheric carbon than a natural forest. At
the same time, many natural ecosystems with low biodiversity
(e.g., tropical freshwater swamps) have a high capacity to fix
carbon.
*The term environmental services is preferred since it embraces the
larger-scale and often more important services, such as the albedo
stabilization supplied by the Amazonia and Zaire forests (14). These
forest regions are too large to conform to the category of ecosystems
as conventionally understood.
 Ecology: Myers
Similarly, the services supplied by one form of biodiversity
in one locality may not necessarily be supplied by a similar form
of biodiversity in another locality. Just because a wetland on
the Louisiana coast performs a particular suite of functions, it
cannot be assumed that a wetland on the Georgia coast will
perform the same functions, still less an inland wetland in
Massachusetts or California, and still less again a montane
wetland in Sweden or a forest wetland in Thailand. Services
tend to be quite site-specific. This makes it much more difficult
for conservation biologists to demonstrate the intrinsic value
of wetlands or any other biotopes.
Biodiversity plays two critical roles. (i) It provides the
biospheric medium for energy and material flows, which in
turn provide ecosystems with their functional properties; and
(ii) it supports and fosters ecosystem resilience (17, 19-24). As
biodiversity is depleted, there is usually a shift and often
(though not always) a decline in the integrity of ecosystem
processes that supply environmental services.
The second caveat is that we should distinguish between
what can be called the ecologist's and the economist's mode of
calculation of values at issue. The first favors estimating
biodiversity's values "in themselves," i.e., the worth of a
biodiversity attribute as manifested by its role in ecosystem
workings (for example, the part played by forest cover in
watershed functions). The second approach prefers to consider
biodiversity's contributions as economic attributes, looking at
the consequences of biodiversity decline for economic activi-
ties (both production and consumption) and assessing the
resultant costs such as prevention of damage, compensatory
alternatives, and substitutes (if any) (2, 11, 25). All this is a
highly anthropocentric approach, and many ecologists con-
sider that it severely underestimates total values at stake.
Both these approaches run the risk of viewing a segment of
biodiversity-whether a species, a population, a gene reser-
voir, a biotope, or a biota-in isolation from its support system.
A mammal species, a butterfly community, a wetland food
web, or a forest ecosystem cannot exist except within the
myriad ecological relationships and ecosystem processes
(moisture supply, nutrient cycling, energy flow, and the like)
of its environs, much if not most of which makes scant if any
direct utilitarian contribution to human welfare. The only
worthwhile approach is to conserve the lot-a strategy to be
followed in fortunate circumstances such as when a keystone
species serve as a flagship species too.
III. Assessment of Select Environmental Services
Environmental services are so abundant and diverse that I
cannot do more here than look at an illustrative selection [for
a comprehensive treatment from an ecological standpoint, see
Ehrlich and Ehrlich (1), and from an economic standpoint, see
Pearce and Moran (26)]. Note that while biodiversity plays a
part in all these services, it may not always play a predominant
or even a primary role, even though it almost always seems to
play a significant role. In Instances 1-4 (below), the key contri-
bution may be supplied by biomass or some other attribute of
biotas and ecosystems (for further elaboration of this key ques-
tion, see Section V below, Biodiversity and Biomass).
1. Climate. Biodiversity helps to maintain the gaseous
composition of the atmosphere and thus to regulate climate. It
also affects climate by cycling vast amounts of water. A leading
example is provided by Amazonia, which contains two-thirds of
all above-ground freshwater on Earth. At least half ofAmazonia's
moisture is retained within the forest ecosystem, being constantly
transpired bya mean
plants before being precipitated back onto the
forest, with recycling time of 5.5 days (27).
There are other biodiversity/rainfall connections. In several
parts of the humid tropics-the Panama Canal Zone, north-
western Costa Rica, southwestern Ivory Coast, montane Tan-
zania, southwestern India, northwestern Peninsular Malaysia,
Proc. Natl. Acad. Sci. USA 93 (1996) 2765
and parts of the Philippines among other areas-rainfall
regimes have been disrupted if not depleted in the wake of
deforestation (28).
2. Biogeochemical Cycles. The Earth's biotas are prime
pumps in the major biogeochemical cycles (29-31). It is
debatable, however, how far this function is impaired by loss
of biodiversity in itself, rather than by loss of vegetation and
other biomass (32, 33).
A notable illustration lies with the carbon cycle and, hence,
with climate change in the form of global warming. Roughly
half of global warming is due to build-up of anthropogenic
carbon dioxide in the global atmosphere. More carbon dioxide
is released than remains in the atmosphere, the rest being
absorbed by the oceans and terrestrial biotas.
While vegetation can serve as a major sink of carbon
dioxide, we do not know how much, nor how far, the function
depends on biodiversity. Preliminary evidence shows that
species-rich ecosystems can often (though not always) con-
sume carbon dioxide at a faster rate than less diverse ecosys-
tems; and in turn this suggests that biodiversity decline may
promote the build-up of carbon dioxide (34). In addition, to
the extent that species-rich ecosystems produce more biomass,
they consume more carbon dioxide, thereby reducing the
build-up of carbon dioxide.
The value of carbon storage in tropical forests as a counter
to global warming is estimated to be in the order of $1000 to
$3500 per hectare per year (35), depending upon the type of
forest and primarily reflecting the amount of biomass in the
forest (rather than the amount of biodiversity in forest com-
munities-though the first is to some extent a function of the
second). The value of the carbon storage service supplied by
Brazilian Amazonia is estimated to be some $46 billion (36).
It has been further estimated that replacing the carbon storage
function of all tropical forests would cost $3.7 trillion (37). But
note a strong qualifier: the forests supply the service by virtue
of their biomass, in which biodiversity appears to play a vital
though not predominant role.
3. Hydrological Functions. Plants play a part in hydrolog-
ical cycles in addition to those cited in Item 1, by controlling
water runoff. Thick and sturdy vegetation permits a slower and
more regulated runoff, allowing water supplies to make a
steadier and more substantive contribution to their ecosys-
tems, instead of quickly running off into streams and rivers-
possibly resulting in flood and drought regimes downstream.
Excessive runoff causes soil erosion in catchment zones and
siltation in valleyland water courses. Siltation of reservoirs
costs the global economy some $6 billion a year in lost
hydropower and irrigation water (37).
In the 120-km2 Bacuit Bay with its 78-km2 drainage basin on
Palawan Island in the Philippines, logging on steep slopes has
increased soil erosion 235 times above that for undisturbed
forest, with a "silt smother" effect for the Bay's coral reef and
its fisheries that reduced commercial revenues by almost half
in the mid-1980s (38). The montane forest of the Rwanda
Volcanoes Park, home to one of the last populations of the
mountain gorilla, covers only about 1% of Rwanda but acts as
the sponge that absorbs and metes out about 10% of agricul-
tural water for that severely overpopulated nation (39). At the
Korup Park in Cameroon, watershed functions (flood preven-
tion, protection of fisheries, and soil conservation) have a net
present value of $85 per hectare (40). Inlevied Java, siltation of
reservoirs, irrigation systems and harbors damage costs
worth $58 million in 1987, equivalent to 0.5% of agricultural
gross domestic product (41).
Consider too the important though little recognized services
performed by wetlands. These services include supply of
freshwater for household needs, sewage treatment, cleansing
of industrial wastes, habitats for commercial and sport fisher-
ies, recreation sites, and storm protection (42). Their economic
values can be sizeable (43). Louisiana wetlands are estimated
2766 Ecology: Myers
to be worth $6000-16,000 per hectare with an 8% discount rate
or $22,500-42,500 per hectare with a 3% discount rate. At the
lowest value, the current annual rate of loss of these wetlands is
levying costs of about $600,000 per km2 peryear, and at the largest
value, levying $4.4 million (late 1980s values). The most valuable
wetlands service by far is storm protection (44). Marshlands near
Boston are valued at $72,000 per hectare per year solely on the
basis of their role in reducing flood damage (45).
Particularly important wetlands are to be found in estuaries.
They feature rapid plant and algal growth that provides the
start of food chains for local fisheries, and they serve as
nurseries for the juvenile stages of many marine fish species.
In the past 50 years, many U.S. estuaries have been severely
damaged by industrial pollution, dumping of untreated resi-
dential sewage, and coastal development. The National Ma-
rine Fisheries Service estimates this damage has cost the
nation more than $200 million per year just in the lost
productive value of commercial fish and sport fish (2).
4. Soil Protection. In similar style, vegetation and to some
extent biodiversity protects soil cover. Soil erosion is a major
problem in many parts of the world since it leads to (i)
significant declines in soil fertility and, thus, in the productivity
of croplands and pastures and (ii) sedimentation of rivers and
other water bodies affecting downstream communities. Four-
fifths of the world's agricultural soils are affected by erosion,
and every year 75 billion tonnes (1 tonne = 1000 kg) of topsoil
are washed or blown away, causing 80,000 km2 to be lost to
agriculture. In the past 200 years, the average topsoil depth in
the United States has declined from 23 cm to 15 cm, costing
the American consumer around $300 per year through loss of
nutrients and water and with total costs to the United States
of $44 billion. Worldwide costs of soil erosion are in the order
of $400 billion per year or equivalent to half of what the world
spends on military activities (46).
5. Crop Pollination. About one-third of the human diet
depends on insect-pollinated vegetables, legumes, and fruits.
Wild bees and honey bees pollinate $30 billion worth of 90 U.S.
crops annually, plus many natural plant species. On a bright
sunny day in upstate New York, bees can pollinate as many as
one trillion blossoms. Honey bee numbers in the United States
have dropped by about 20% during the period from mid-1990
to mid-1994, due to the introduction of two alien parasitic
mites. As a result, almond growers in California, with a crop
worth $800 million a year, have had to import bees from as far
away as Florida and South Dakota. Pollination is a service for
which there is no technological substitute (47).
6. Pest Control. Around 35% of the world's crop produc-
tion is lost to pests, of which there are at least 67,000
recognized species. Only about 300 species have been targeted
by biological controls, and of these 120 species have been
success stories. So there is much scope to draw on the vast stock
of natural controls "out there" in the form of predators and
parasites, plus host plant resistance (17, 48, 49).
7. Ecotourism. Biodiversity plays a vital part in the fast-
growing sector of ecotourism. Each year people taking nature-
related trips contribute to the national incomes of countries
concerned a sum estimated to be at least $500 billion, perhaps
twice as much (50, 51). Much of the enjoyment of these
ecotourists reflects the biodiversity they encounter.
In the late 1970s, a single lion in Kenya's Amboseli Park
earned $27,000 per year in tourist revenues, while an elephant
herd earned $610,000 per year (52). In 1994, whale watching
in 65 countries and dependent territories attracted 5.4 million
viewers and generated tourism revenues of $504 million, with
annual rates of increase of more than 10% and almost 17%,
respectively. A pod of 16 Bryde's whales at Ogata in Japan
would, according to very conservative estimates, earn at least
$41 million from whale watchers over the next 15 years (and
be left alive), whereas if killed (as a one-shot affair) they would
generate only $4.3 million (53). In 1970, ecotourism in Costa
Proc. Natl. Acad. Sci. USA 93 (1996)
Rica's Monteverde Cloud Forest Reserve generated revenues
of $4.5 million, or $1250 per hectare-to be compared with
$30-100 per hectare for land outside the reserve (54). Florida's
coral reefs are estimated to be worth $1.6 billion a year in
tourism revenues (55).
8. Miscellaneous Services. Now for a brief selection of
some further services: The larvae of certain aquatic flies-
notably mayflies, stone flies, caddis flies, and true flies-can
be used to identify point sources of chemical contaminants in
water bodies, especially with respect to molybdenum, manga-
nese, and copper (56). Other species, such as earthworms and
certain fish, birds, and mammals, serve as biological monitors
of various kinds of widespread pollution. A number of wild
plants, for instance the water hyacinth, act as first-rate depol-
luting agents in sewage lagoons. A few plant species can even
register radiation, some of them more sensitively than a
dosimeter (57).
A number of tree species (beech, elm, oak, sycamore, willow,
and elder) in cities serve to clean up pollution, notably sulfur
dioxide (58). Trees also act as air coolants. A 20-m shade tree
can mitigate 900,000 BTUs (1 BTU = 1060 J) of heat, worth
three tonnes of air conditioning a day at a cost of $20 in the
United States (early 1980s value) (59).
Certain animals, such as dogs, cats, horses, chimpanzees,
and snakes, appear able to anticipate even slight earth tremors
and, thus, to warn of impending earthquakes (60).
Many species act as research models. The woodpecker, with
a neck built to withstand severe whiplash, has offered a
blueprint for crash helmets. A species of chalcid wasp with
unusual capacity for hovering has aided with the design of an
improved helicopter. Squids, with nerve fibers 1000 times
larger in cross-section than human nerve fibers, supply neu-
roscientists with crucial insights into the human nervous
system.
For a lengthy listing of such miscellaneous items, see Myers
(3).
IV. Ecosystem Resilience
Many of the services listed above are closely associated with
the phenomenon of ecosystem resilience. If this resilience
declines, the services can generally be expected to decline, too.
This aspect is so important that, while it can be characterized
as a service (or rather, as a kind of super-service), it warrants
treatment on its own.
Resilience can be defined as the ability of ecosystems to
resist stresses and shocks, to absorb disturbance, and to recover
from disruptive change (many of these perturbations being due
to human activity and especially economic activity) (19, 24, 61,
62). Or, to express the concept more formally, it connotes an
equilibrium-theory idea to the effect that ecosystems with their
cybernetic mechanisms display homeostatic attributes that
allow them to maintain function in the face of stress-induced
structural changes (15, 63). How far is ecosystem resilience
dependent on biodiversity? If there is indeed a directly caus-
ative connection, this may turn out to be the number one
service supplied by biodiversity insofar as all other services
appear to depend on it to some degree (19, 64, 65).
There is some evidence that biodiversity can make an
important contribution to ecosystem resilience (66-68). At the
same time, there is much uncertainty about several associated
factors (69): the range of species composition within which
ecosystems and communities function (70); the part played by
species richness (only one aspect of biodiversity) in ecosystem
attributes such as trophic structures and successional stages
(62); the contribution of dominant species such as keystone
mutualists and critical-link species (71); the link between
biodiversity and ecosystem scale (13); and the relationships
among biodiversity, biomass, and ecosystem productivity (19)
(for more on this last point, see Part V below). Moreover, each
 Ecology: Myers
of these factors may operate differently when an ecosystem is
at equilibrium or in transition (62). All this means that the
environmental services supplied by biodiversity may be quite
wide in scope, while localized in scale and particularized in
effect (23, 72).
There is much controversy about the biodiversity/resilience
relationship. In certain circumstances, biodiversity can en-
hance ecosystem performance overall (19, 62). For instance,
more diverse plant communities tend to reveal greater primary
productivity under conditions of laboratory experiment (34,
73); and the primary productivity of species-rich plant com-
munities in grasslands is more resistant to, and recovers more
fully from, major stresses such as droughts (24, 74-76). Then
there is uncertainty about threshold effects (except that they
are specially significant). At what point of biodiversity decline
do ecosystems start to lose the self-organizing capacity and,
hence, the resilience that apparently enables them to provide
certain environmental services (67)? In terms of net primary
productivity, evidence suggests that above a threshold number
of species, there is no gain in function (16).
The biodiversity contribution to resilience extends of course
to subunits of species. Populations differ in their genetic
structure by virtue of their adaptation to environmental con-
ditions and random chance. The genetic variability represented
by geographically disparate populations helps assure the ability
of an entire species to respond evolutionarily to environmental
change (13, 77, 78). If, for example, there is rapid climatic
change (as is likely to ensue through global warming), a species
with many populations is more likely to include members
genetically adapted to the new conditions than a species with
a single population.
This behoves us to expand our purview of the mass extinc-
tion crisis underway and to consider a crunch question. Sup-
pose, as is entirely likely, that within the coming few decades
we lose 50% of all species and 90% of populations of surviving
species. Which will entail the greatest repercussions for eco-
system resilience in a world undergoing environmental up-
heaval of altogether unprecedented scope and scale? This is a
vital issue for conservation biologists-also political leaders,
policy makers, the general public, and indeed anybody con-
cerned about the future habitability of the biosphere.
Herein too lies the question of species redundancy. This
postulates that many if not most species are not required for
ordinary ecosystem functioning. As noted, ecosystem pro-
cesses often appear to be quite resilient to biodiversity decline:
they can keep on supplying environmental services after losing
a good number of species and large numbers of populations
(74). Plainly, then, there is much redundancy built into nature.
Britain has lost the bear, the wolf, and other top carnivores,
plus many herbivores and perhaps detritivores, with little if any
apparent harm to its ecosystems (albeit thanks in many in-
stances to compensatory management such as sport hunting
and culling of deer). North America, Madagascar, and Aus-
tralia have lost a large share of their vertebrate megafauna
within the recent past, yet there is scant evidence of profound
or pervasive ecosystem decay (but see ref. 79).
It is incorrect, then, to say that each species has its essential
part to play in ecosystems, let alone that it is a mainstay of
stability or resilience. It is also incorrect to say that we can lose
lots of species with impunity. A cut-off stage would (eventu-
ally) arrive when there would be simply too few species to
maintain basic ecosystem functions. Where is the "grey" zone
where biodiversity decline starts to approach the threshold of
irreversible ecosystem injury? Scientists have all too little idea,
and so they would do well to recall the rivet popping analogy
(1). Similarly pertinent is the notion that redundancy itself may
well have a functional value for ecosystems, as a kind of
"nature's insurance"-a benefit that generally becomes oper-
ative only within extended time frames (23, 24, 68). To this
extent, we may eventually find that biodiversity contributes an
Proc. Natl. Acad. Sci. USA 93 (1996) 2767
environmental service of semiabsolute value in the sense of
reducing severe risk but that it plays only a relatively significant
role in supplying the many other services listed.
In conclusion to this review of ecosystem resilience and of
biodiversity's part in it, recall that the issue is so beset with
uncertainty of multiple sorts that we shall never be scientifi-
cally assured as to how far biodiversity limits can be pressed
before unacceptable risks are encountered. Final knowledge
comes only with a post mortem. Note the warning of that
biodiversity doyen, Edward O. Wilson (80):
"If enough species are extinguished, will ecosystems
collapse and will the extinction of most other species
follow soon afterwards? The only answer anyone can
give is: Possibly. By the time we find out, however, it
might be too late. One planet, one experiment."
V. Biodiversity and Biomass
Much of Part IV has brought up a basic issue: environmental
services often appear to depend not only or not so much on
biodiversity as on biomass. When a patch of natural forest in
the humid tropics is eliminated in favor of a commercial pine
plantation or even a tea crop (dozens of plant species replaced
by one), the new vegetation can supply certain of the same
ecological functions, notably protection of soil cover and
hydrologial systems (81). Similarly, it is not only biodiversity
that enables plants to exploit energy from the sun. Photosyn-
thesis can often be generated most productively (though
perhaps with less long-term stability) by a monoculture of, e.g.,
sugarcane. So it is important not to confuse biodiversity with
biomass-or, for that matter, associated factors such as com-
munity make-up and vegetation structure.
Plants cycle moisture from the soil. A single rainforest tree
can, during a lifetime of 100 years, return at least 10 million
liters of water to the atmosphere (1). But a succession of 12,000
corn stalks occupying 0.1 hectare (roughly the same area as
taken up by a rainforest tree) for a few months each year
would, in the case of the United States and during the same
century, transfer 0.5 million liters per year and 50 million liters
in 100 years-though the corn would need massive inputs of
synthetic fertilizer and other agronomic inputs to do it (46).
Similarly, a carbon sink can be maintained by a tract of
rainforest or a plantation of eucalyptus trees-though a plan-
tation would probably provide less cycling of minerals and
other soil nutrients, be more vulnerable to pest outbreaks, and
supply next to nothing in the way of "genetic library" services.
The estimated 20,000 species of ants number somewhere
between one trillion and 10 trillion individuals, with a biomass
as much as all humankind; in certain localities they can make
up 25% of the animal biomass, and in sectors of the Amazonia
forest they constitute more than four times the biomass of all
land vertebrates combined (82). We can still ask, however,
whether ants' multifarious activities could not be performed
more or less as well with an equal amount of biomass con-
taining far fewer species.
VI. Some Economic Dimensions
It is the aim of this paper to demonstrate the scope and scale
of environmental services and their values, rather than to
engage in a comprehensive assessment of their values in
economic terms. Of course it helps to have some idea of how
far the economic values are significant, and so the paper
presents a few illustrative instances of values in question.
More revealing, however, is an indication-however pre-
liminary and exploratory-of the economic values overall
implicit in the environmental services supplied by some par-
ticular ecosystem or region. Note, then, that the annual value
of nonmarketed environmental services provided primarily by
2768 Ecology: Myers
wetlands, forests, and agricultural areas in the state of Georgia
have been estimated to be worth $2.6 billion in 1982 dollars-a
sum to be compared with the annual value of the state's
marketed agricultural products, $2.8 billion, and marketed
timber products, $4.5 billion (83). In the state of Oregon,
environmental services in the form of amenity alone are
estimated to be worth at least $500 per citizen per year (84).
On a larger scale, consider the cost of Biosphere 2, being the
man-made technosphere in the Arizona desert that (margin-
ally) regulated life-support systems for eight Biospherians over
2 years: about $150 million, or $9 million per person per year.
These same services are provided to the rest of us by natural
processes, at no cost. But if we were charged at the rate levied
by Biosphere 2, the total bill for all Earthospherians would
come to $3 quintillion for the current generation alone (85).
VII. Conclusion
First, this paper demonstrates that (i) the environmental
services of biodiversity are certainly significant, probably much
more so than the direct benefits of biodiversity in the form of
material goods; and (ii) all too little is known about the nature,
scope, and scale of these services, whether in environmental or
economic senses. This places a premium on research to
increase our understanding-a challenge made all the more
pressing by the expansion of the human niche and all that
entails for progressively increasing pressures on biodiversity's
habitats and life-support systems.
It might not be of much profit, however, to engage in more,
and more detailed, documentation of the services, even though
no more than a start on gathering data and other forms of
information has been made. The critical track ahead lies not so
much with knowledge as with understanding. A far greater
analysis of basic key questions is needed, such as: how does
biodiversity generate environmental services; how much biodi-
versity is needed to do the job; and how far does the relation-
ship depend on local circumstances, especially site conditions
(which may change over time)? On top of these questions and
others already recognized, there are surely other vital ques-
tions that have not even been identified and defined.
Herein lies the biggest challenge of all, to determine a
comprehensive answer to the point posed at the start of this
article, What is biodiversity good for? At present rates of
research and analysis, responses to that question may eventu-
ally be found only by discovering what has been lost after much
biodiversity together with its environmental services has been
eliminated.
A second conclusion is that conservation biologists should
feel more inclined to simply reject the question, What is
biodiversity good for? There will not be anywhere near a
sufficient answer within a time frame to conclusively persuade
political leaders, policy makers, and the public (let alone the
professional skeptics). Rather, the uniqueness and irrevers-
ibility arguments should be invoked and thus the burden of
proof should be thrown on the doubters, requiring them to
demonstrate that biodiversity is generally worth so little that it
can be dispensed with if human welfare demands as much
through, e.g., agricultural encroachment on wildland habitats.
True, there is vast uncertainty about what biodiversity con-
tributes to the human cause. But due to the asymmetry of
evaluation, the doubters are effectively saying they are com-
pletely certain that we, and our descendants for millions of
years (until evolution restores the loss), can manage well
enough without large quantities of biodiversity.
I assert, above all, that biodiversity conservation is comple-
mentary to, rather than competitive with, other pursuits of
human well-being. The time has come when biodiversity
cannot be safeguarded primarily in protected areas. For one
thing, there is not nearly enough of them in the right places,
and most of them are too small-and there is poor chance that
Proc. Natl. Acad. Sci. USA 93 (1996)
many more can be established in an increasingly crowded
world. For another thing, one-third of protected areas in the
tropics (the most vital zone from the biodiversity standpoint)
are already being encroached upon by expanding agriculture,
and this trend is likely to accelerate given the burgeoning
numbers of land-hungry peasants. For still another and yet
more significant thing, several of the best managed parks and
reserves are being overtaken by acid rain; similarly, no pro-
tected area can ever be shielded from UV-B radiation and
global warming. Within a few decades, indeed, there may be no
more protected areas [to cite McNeely (86)]:
"either because they have been over-taken by land-
hungry peasants or grandscale pollution, or because we
have finally found a way to manage all our landscapes
that the needs of biodiversity are taken care of auto-
matically."
This all means that biodiversity can ultimately be saved only
by saving the biosphere as well. Thus the following things must
be undertaken on all kinds of other good grounds: stem acid
rain, push back the deserts, replant the forests, restore topsoil,
reverse ozone-layer depletion, stabilize climate, etc. (also of
course halt population growth, reduce overconsumption, cut
back on global inequities, etc.). In this writer's view, it is far
more important to focus on ways to meet these imperatives
than to engage in finer-grain assessment of environmental
services.
Emphatic thanks to David Duthie and Stuart Pimm for their
unusually helpful comments on an early draft of this paper. This article
has been written with funding through the Pew Fellows Programme.
1. Ehrlich, P. R. & Ehrlich, A. H. (1992) Ambio 21, 219-226.
2. McNeely, J. A. (1988) Economics and Biological Conservation:
Developing and Using Economic Incentives to Conserve Biological
Resources (Int. Union for Conservation of Nature and Natural
Resources, Gland, Switzerland).
3. Myers, N. (1983) A Wealth of Wild Species (Westview, Boulder,
CO).
4. Oldfield, M. L. (1989) The Value of Conserving Genetic Resources
(Sinauer, Sunderland, MA).
5. Westman, W. E. (1977) Science 197, 960-964.
6. Bishop, R. C. (1987) in Valuing Wildlife: Economic and Social
Perspectives, eds. Decker, D. J. & Goff, D. R. (Westview, Boul-
der, CO), pp. 24-34.
7. di Castri, F. & Younis, T. (1990) Ecosystem Function of Biodi-
versity (Int. Union Biol. Sci., Paris), Biol. Int. Special Issue 22.
8. Green, C. H. & Tunstall, S. M. (1991) J. Environ. Manage. 33,
123-141.
9. Risser, P. G. (1995) Conserv. Biol. 9, 742-746.
10. Barbier, E. B., Burgess, J. C. & Folke, C. (1994) Paradise Lost?
The Ecological Economics of Biodiversity (Earthscan, London).
11. Brown, K., Pearce, D., Perrings, C. & Swanson, T. (1993)
Economics and the Conservation of Global Biological Diversity
(Global Environment Facility, The World Bank, Washington,
DC).
12. Ehrlich, P. R. & Wilson, E. 0. (1991) Science 253, 758-762.
13. Ehrlich, P. R. & Daily, G. C. (1993) Ambio 22, 64-68.
14. Gash, J. H. C. & Shuttleworth, W. J. (1992) in Tropical Forests
and Climate, ed. Myers, N. (Kluwer Academic, Dordrecht, The
Netherlands), pp. 123-134.
15. Cairns, J. & Pratt, J. R. (1995) in Evaluating and Monitoring the
Health of Large-Scale Ecosystems, eds. Rapport, D. J., Gaudet,
C. L. & Calow, P. (Springer, Berlin), pp. 63-76.
16. Perrings, C. (1987) Economy and Environment: A Theoretical
Essay on the Interdependence of Economic and Environmental
Systems (Cambridge Univ. Press, Cambridge, U.K.).
17. Baskin, Y. (1994) BioScience 44, 657-666.
18. Cairns, J. & Niederlehner, B. R. (1994) Environ. Health Perspect.
102, 936-939.
19. Schulze, E.-D. & Mooney, H. A., eds. (1994) Biodiversity and
Ecosystem Function (Springer, New York), Study Ed.
 Ecology: Myers
20. Ehrlich, P. R. & Roughgarden, J. (1987) The Science of Ecology
(Macmillan, New York).
21. Folke, C. & Kaberger, T., eds. (1991) Linking the Natural
Environment and the Economy (Kluwer Academic, Dordrecht,
The Netherlands).
22. Holling, C. S. (1986) in Sustainable Development of the Biosphere,
eds. Clark, W. C. & Munn, R. E. (Cambridge Univ. Press,
Cambridge, U.K.).
23. Perrings, C. (1995) in The Economics and Ecology of Biodiversity
Decline: The Forces Driving Global Change, ed. Swanson, T. M.
(Cambridge Univ. Press, Cambridge, U.K.), pp. 69-77.
24. Walker, B. H. (1992) Conserv. Biol. 6, 18-23.
25. Aylward, B. & Barbier, E. B. (1992) Biodiversity Conserv. 1,
34-50.
26. Pearce, D. & Moran, D. (1994) The Economic Value of Biodi-
versity (Earthscan, London).
27. Salati, E. & Nobre, C. A. (1992) in Tropical Forests and Climate,
ed. Myers, N. (Kluwer Academic, Dordrecht, The Netherlands),
pp. 177-196.
28. Meher-Homji, V. M. (1992) in Tropical Forests and Climate, ed.
Myers, N. (Kluwer Academic, Dordrecht, The Netherlands), pp.
163-173.
29. Odum, E. P. (1993) Ecology and Our Endangered Life-Support
Systems (Sinauer, Sunderland, MA).
30. Ricklefs, R. E. (1987) Science 235, 167-171.
31. Schlesinger, W. H. (1991) Biogeochemistry: An Analysis of Global
Change (Academic, San Diego).
32. Kareiva, P. M., Kingsolver, J. G. & Huey, R. B., eds. (1993) Biotic
Interactions and Global Change (Sinaeur, Sunderland, MA).
33. Vitousek, P. M. & Hooper, D. U. (1994) in Biodiversity and
Ecosystem Function, eds. Schulze, E.-D. & Mooney, H. A.
(Springer, New York), pp. 3-14.
34. Naeem, S., Tompson, L.J., Lawler, S. P., Lawton, J. H. &
Woodfin, R. M. (1994) Nature (London) 368, 734-737.
35. Pearce, D. & Moran, D. (1995) The Economic Consequences of
Biodiversity Loss (Centre for Social and Economic Research on
the Global Environment, University College, London).
36. Gutierrez, B. & Pearce, D. W. (1992) Estimating the Environ-
mental Benefits of the Amazon Forest: An International Evaluation
Exercise (Centre for Social and Economic Research on the Global
Environment, University College, London).
37. Panayotou, T. & Ashton, P. S. (1992) Not By Timber Alone
(Earthscan, London).
38. Hodgson, G. & Dixon, J. A. (1988) Logging Versus Fisheries and
Tourism in Palawan (East-West Environment and Policy Inst.
East-West Center, Honolulu).
39. McNeely, J. A. & Miller, K. R. (1984) National Parks, Conserva-
tion and Development: The Role of Protected Areas in Sustaining
Society (Smithsonian Inst. Press, Washington, DC).
40. Ruitenbeek, J. (1989) Social Cost-Benefit Analysis of the Korup
Project, Cameroon (World Wide Fund for Nature-U.K., Godalm-
ing, U.K.).
41. Magrath, W. & Arens, P. (1989) The Costs of Soil Erosion on Java:
A Natural Resource Accounting Approach (The World Bank,
Washington, DC).
42. Mitsch, W. J. & Gosselink, J. G. (1993) Wetlands (Reinhold, New
York), 2nd Ed.
43. Folke, C. (1991) in Linking the Natural Environment and the
Economy, eds. Folke, C. & Kaberger, T. (Kluwer, Dordrecht, The
Netherlands), pp. 141-171.
44. Costanza, R., Farber, S. C. & Maxwell, J. (1989) Ecol. Econ. 1,
335-361.
45. Hair, J. D. (1988) The Economics of Conserving Wetlands: A
Widening Circle (Natl. Wildlife Fed., Washington, DC).
46. Pimentel, D., Harvey, C., Resosudarmo, P., Sinclair, K., Kurz, D.,
McNair, M., Crist, S., Shpritz, L., Fitton, L., Saffouri, R. & Blair,
R. (1995) Science 267, 1117-1122.
47. Pimentel, D., Stachow, U., Takacs, D. A., Brubaker, H. W.,
Dumas, A. R., Meaney, J. J., O'Neil, J. A. S., Onsi, D. E. &
Corzilius, D. B. (1992) BioScience 42, 354-362.
Proc. Natl. Acad. Sci. USA 93 (1996) 2769
48. Pimentel, D. (1991) Crop Prot. 10, 243-253.
49. Gillis, A. M. (1992) BioScience 42, 482-485.
50. Eagles, P. F. J., Buse, S. D. & Hvengaard, G. T., eds. (1993) The
Ecotourism Society Annotated Bibliography (Ecotourism Soc.,
Bennington, VT).
51. Whelan, T., ed. (1991) Nature Tourism: Managing for the Envi-
ronment (Island Press, Washington, DC).
52. Western, D. & Henry, W. (1979) BioScience 29, 414-418.
53. Hoyt, E. (1995) The Worldwide Value and Extent of Whale
Watching, 1995 (Whale and Dolphin Conserv. Soc., Bath, U.K.).
54. Tobias, D. & Mendelsohn, R. (1991) Ambio 20, 91-93.
55. Adams, J. (1995) Ecotourism: Conservation Tool or Threat?
(World Wildlife Fund-US, Washington, DC).
56. Root, M. (1990) BioScience 40, 83-86.
57. Mansfield, T. A., ed. (1976) Effects of Air Pollutants on Plants
(Cambridge Univ. Press, Cambridge, U.K.).
58. Roberts, B. R. (1979) Trees as Biological Filters (U.S. Dept.
Agriculture, Delaware, OH).
59. Dove, M. R. (1982) cited in Cairns, J. & Niederlehner, B. R.
(1994) Environ. Health Perspect. 102, 936-939.
60. Kerr, R. A. (1980) Science 208, 695-696.
61. Ehrlich, P. R. (1986) The Machinery of Nature (Simon & Schuster,
New York).
62. Pimm, S. L. (1991) The Balance of Nature? Ecological Issues in the
Conservation of Species and Communities (Univ. Chicago Press,
Chicago).
63. Odum, E. P. (1985) BioScience 35, 419-422.
64. Perrings, C., Maler, K.-G., Folke, C., Holling, C. S. & Jansson,
B. O., eds. (1995b) Biodiversity Loss: Ecological and Economic
Issues (Cambridge Univ. Press, Cambridge, U.K.).
65. Ricklefs, R. E. & Schluter, D., eds. (1993) Species Diversity in
Ecological Communities (Univ. Chicago Press, Chicago).
66. Jones, C. G. & Lawton, J. H., eds. (1995) Linking Species and
Ecosystems (Chapman & Hall, London).
67. Pimm, S. L. (1994) in Biodiversity and Ecosystem Function, eds.
Schulze, E.-D. & Mooney, H. A. (Springer, New York), pp.
347-360.
68. Walker, B. (1995) Conserv. Biol. 9, 747-752.
69. Perrings, C. & Opschoor, H. (1994) Environ. Resource Econ. 4,
1-11.
70. Solbrig, O. T. (1991) Environment 33, 10-23.
71. Mills, L. S., Soule, M. E. & Doak, D. F. (1993) BioScience 43,
219-224.
72. Common, M. & Perrings, C. (1992) Ecol. Econ. 6, 9-15.
73. Kareiva, P. (1994) Nature (London) 368, 686-687.
74. Lawton, J. H. & Brown, V. K. (1994) in Biodiversity and Ecosys-
tem Function, eds. Schulze, E.-D. & Mooney, H. A. (Springer,
New York), pp. 255-270.
75. Tilman, D. & Downing, J. A. (1994) Nature (London) 376,
363-365.
76. Walker, B. H. (1993) Ambio 22, 80-87.
77. Caro, T. M. & Laurenson, M. K. (1994) Science 263, 485-486.
78. Rawls, K., Ballou, J. D. & Brownwell, R. L. (1993) Biol. Conserv.
25, 209-215.
79. Owen-Smith, N. (1988) Megaherbivores: The Influence of Very
Large Body Size on Ecology (Cambridge Univ. Press, Cambridge,
U.K.).
80. Wilson, E. 0. (1992) The Diversity of Life (Harvard Univ. Press,
Cambridge, MA).
81. Myers, N. (1992) The Primary Source (Norton, New York).
82. Wilson, E. 0. (1995) Naturalist (Allen Lane, London).
83. Turner, M. G., Odum, E. P., Costanza, R. & Springer, T. M.
(1988) Environ. Manage. 12, 209-217.
84. Durning, A. T. (1992) How Much is Enough? (Worldwatch Inst.,
Washington, DC).
85. Avise, J. C. (1994) Conserv. Biol. 8, 327-329.
86. McNeely, J. A. (1990) Environment (Jan/Feb), 16-20 and 36-41.